This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Based on the processes, activities and materials inventory at the Schwarz Pharma Ltd facility it has been established that the site is a lower tier site as establishment under the terms of articles of the European Communities (Control of Major Accident Hazards Involving Dangerous Substances) Regulations, 2000 (SI No. 476 of 2000)..The Health and Safety Authority and Clare County Council where notified of the plant status as a lower tier site in 2001, the letter of notification and the calculation which formed the basis of the site classification is presented in Appendix B.9.1.
2. PROCESS SAFETY MANAGEMENT SYSTEM
In order to ensure the ongoing control of major accident hazards at the Schwarz facility the company has implemented a process safety management system (SOP 601.032). This SOP is contained in Attachment J. The purpose of this procedure is to describe the methodology in use at Schwarz Pharma Ltd to ensure that
l I* The safety of all processes is assessed prior to their introduction or modification
l The necessary controls eliminate or minimise any associated risk.
l The necessary Emergency measures are put in place to mitigate the consequences of an associated emergency situation should it arise.
This system includes procedures relating to the following:
Chemical Reaction Hazard Studies SOP 940.001
Equipment Proposal Report - SOP 211.011
Preliminary Hazard Assessment - SOP 601.031
Equipment Evaluation & Selection -SOP 301.003
HAZOP Procedure - SOP 311.001,
Basis of Process Safety Report - SOP 311.002
Basis of Safety & Environmental Protection - SOP 601.030.
The company has implemented a number of arrangements to ensure compliance with these regulations . In May 2002 a Major Hazard Study, Seveso II Directive Major Hazard Study -SIFA Lid, Shannon Co. Clare, was performed over a twelve-month period at the facility. This study is presented in Appendix B.9.3. The initial objective of this study was to formally identify the major hazard scenarios at the facility using a Preliminary Hazard Analysis (PHA) technique. The technique is detailed in SOP 601.031 is contained in Attachment J. This study has been included as a source of supporting information. On foot of this PHA a representative number of hazard scenarios were selected and were subjected to further analysis in the form of a quantified risk analysis with the aim of arriving at consequence and risk estimates. This quantified risk analysis studied the consequences of airborne releases to atmosphere associated with a number of scenarios. Scenarios included:
. A catastrophic release of nitric acid;
. Toxic gas release effects following a release of nitrogen dioxide;
Notification to Health and Safety Authority and Seveso II Calculation
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:07
SHANNON INDUSTRIALESTATE, co. CIARE. IRELAND.
TEL: +353-@)61-714100 FAX: +353-(0161-714101
Process Industries Unit, I IeaItl~ & Safely Authority, IO Hogan Place, Dublin 2.
Eurcqwan Communities (Control of Major Accident Hazards Involving Uangwous Sulv4tanecs) Regulations, St No. 476 of 2000.
IJncler the rcquirerncnts of the above Statutory Instrument (& under Regulation II - Notif&ion uf I-ktablishmcnts) S[FA I,imiled her&y give notice to the Health and Safety Authority hat this site is subjxt to these Regulations.
The atlachcd folder contains the information as required to be submitted, and as set out in the Third Schedule ofthe Regulations.
This Notikat ion is also being submitted to Ciarc County Council.
‘This site is currently cn~egorised by ‘Articles 6 and 7’ (or Lower ‘I’icr) under the rcquiremcnts o,f the Dircctivc as referred to in the Regulations.
At a future date the site may also lxcomc subject to the rcquircments of Article 9 of the Directive. An application to the HSA will be made in advance of this change and in accordance with the requirements ofthe Regulations.
WA Limited are aIso giving notice IV Glare County CounciI of these potential fixture site developments in the interest of the proper planning and dcvclopmenl of the site and ofthe neighbouring areas in accordance wjth hrticle 12 of the Directive.
1.1 This report forms part of a site evaluation of the major hazard scenarios associated with the chemical manufacturing facility operated by SIFA Ltd located at Shannon, Co. Clare.
1.2 The overpressure effects of a catastrophic vessel rupture, the blast effects from a detonation of process material, and the exposure effects to toxic gas vapours associated with the Nitration Plant are assessed through this report.
Consequence and risk estimates are arrived at using the technique of Quantified Risk Analysis (QRA).
1.3 Individual risk profiles are plotted for the Nitration Plant and the immediate area at SIFA for a number of scenarios.
For acceptability these estimates are compared to industry best practice standards.
1.4 The”results show that the risk profiles of interest arising from toxic gas exposure (namely Nitric Acid and NOx) from a major hazard scenario associated with the Nitration Plant extend to a distance of 700m from the Nitration plant.
Also the risk profiles from the explosion scenarios within the Nitration Plant extend to a distance of 50m from the plant.
The significance of the risk profiles is outlined and explained in Section 7.0.
1.5 At the time of compiling this report a review (and ongoing development) of the process safety management system within SIFA was being undertaken.
Major hazard identification studies conducted as part of this report highlighted a number of safety and environmental issues where further information is required to be compiled and where information management and control is essential to ensuring that the hazards and risk of a major hazard scenario are minimised.
Regular reviews should be conducted until all of the issues identified have been addressed.
Page 3
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:08
SIFA Ltcj Shannon, Co. Ciare.
Quantijied Risk Assessment -Nitration Plant - May 2002
2.0 INTRODUCTION
A revised EC Directive applicable to major hazard industrial and storage sites was implemented into Irish national legislation in December 2000. The Directive 96/82 for the control of Major Accident Hazards generically referred to as the Seveso II Directive (or COMAH) focuses on the prevention of major accidents involving the release of hazardous substances.
The Directive applies to companies who store or process hazardous materials in quantities equal to or in excess of specified thresholds.
A fundamental aspect to the Directive is the requirement for companies to formally identify and assess their site major hazards. The assessment is required to be formal and systematic of the controls in place both for the prevention of incidents and for the mitigation of their consequences.
To assist SIFA in complying with the requirements of the regulations a Quantified Risk Assessment of the major hazard scenarios associated with the Nitration Plant is conducted through thii report.
2.1 Major Hazard Identification
For the purposes of conducting this QRA study it was necessary to identify and categorise the major hazards arising from the processing and storage of hazardous materials within SIFA. To assist in this exercise the technique of Preliminary Hazard Analysis was employed and applied to all areas of the SIFA site.
Preliminary Hazard Analysis is typically used to identify and categorise major site processing and storage hazards involving hazardous materials. It is normally used as a precursor to further analysis of major hazards using the technique of QRA.
A summary of the major hazard scenarios identified for the Nitration Plant is given in Appendix 1.
A representative set of the Nitration Plant major hazard scenarios are quantitatively assessed through the following sections.
2.2 Major Hazards associated with the Nitration Plant
Two active pharmaceutical ingredients namely Isosorbide-5-Mononitrate (5-ISM) and Isosorbide Dinitrate (ISDN) are manufactured in the Nitration Plant.
Production commenced in this purpose built plant in 2001.
The process steps involved in the manufacture of these products have highly hazardous characteristics, namely:
l ISDN has properties similar to that of the commercial explosive TNT.
l At elevated temperatures (i.e. > 12O’C) the manufacturing steps are subject to rapid thermal decomposition at rates where the resulting pressure rise cannot be vented by conventional pressure relief devices. Potentially resulting in catastrophic vessel failure.
l The decomposition of ISDN and 5-ISM results in the formation of highly toxic fumes of NOx.
l A raw material in the manufacturing process is Nitric Acid (98%). The acid is a strong corrosive irritant to the skin, eyes, and mucous membranes.
3.0 QUiNTIFED RISK ASSESSMENT METHODOLOGY AND SCOPE
Chemical process quantitative risk analysis is a methodology designed to help evaluate process safety in the chemical industry. QRA studies provide a quantitative method of evaluating risks and identifying aspects for effective risk reduction.
The basis of QRA is to identify major incident scenarios and evaluate the risk by defining the probability of failure, the probability of various consequences and the potential impact of those consequences.
3.1 Purpose of Study
The purpose of this study is to:
l Compile individual risk proJles associated with the following major hazard scenarios based on the risk of fatality occurring.
l Analyse the profiles and determine if the risks are within acceptable limits when assessed against industry best practices. Where appropriate recommend suitable risk reduction measures,
Page 5
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:08
SIFA Ltd Shannon, Co. Glare.
QuantiJied Risk Assessment -Nitration PIant - May 2002
3.2 Nitration Plant Major Hazard Scenarios
Scenario la: A catastrophic leak fi-om Nitric Acid storage tank TA-640 1 into the tank bund.
Tank TA-6401 is a 20m3 glass lined steel storage tank containing 30,OOOkgs of 96% Nitric Acid.
Scenario lb: A fracture on the 50mm Nitric Acid transfer line from a road tanker to tank TA-6401.
Nitric acid toad tankers contain 20,000 kgs of Nitric acid are delivered to the site approximately 10 times per year, Offloading takes approximately 1 hour.
Scenario lc: A fracture on the 40mm Nitric Acid transfer line from tank TA-6401 to the Nitration Plant.
Nitric acid is pumped from TA-6401 to the Nitration plant via transfer pump PU-6401 at a rate of 5m3&.
The duration of this release is taken to be 10 minutes before intervention to stop Nitric Acid transfer is taken.
Scenario 2a: A catastrophic failure of RE-6101 during the Nitration reaction in the 5-KM process.
Catastrophic failure can occur as a result of a high concentration of Acetyl Nitrate formation leading to a rapid thermal decomposition of the solution within the reactor. The rate of pressure rise potentially resulting in catastrophic vessel failure.
W-6101 is a 2,300 litre vessel with a design pressure of -1 to 3 barg.
Scenario 2b: The emission of NOx fumes as a result of the catastrophic failure of RE-6 10 1.
40,000 litres of NOx is potentially generated and lost instantaneously due to the thermal decomposition of the reaction solution within RE-6 101.
For this scenario the NOx generated is assumed to be Nitrogen Dioxide (NOz).
Scenario 3: The detonation of a batch of 250kgs of pure ISDN within tray drier DR-634 1.
ISDN is similar in its properties to TNT and for this scenario the properties are regarded as being equal for the purposes of calculating the blast effects arising
Page 6
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:08
SIFA LtcJ Shannon, Co. Glare.
Quant@ed Risk Assessment - Nitration Plant - May 2002
3.3 Assumptions Made
In selecting the above major hazard scenarios the following assumptions were made.
3.3.1 For Scenarios 2a and 2b reactor Re-6101 was chosen, as this vessel has the largest volume of material subject to thermal decomposition and consequently the largest volume of permanent gas capable of being generated (i.e. NOx).
3.3.2 Single vessel catastrophic failure is only considered i.e. the rupture of Re-6101 initiating a similar scenario within adjacent vessels is not considered to be realistic (i.e. a domino effect).
3.3.3 Thermal decomposition scenarios involving vessels containing flammable solvent have not being considered as the maximum water temperature capable of being supplied to the jackets of process vessels within the Nitration Plant is 75OC.
The jacket temperature required for thermal decomposition is not considered to be attainable (i.e. 120°C);
It is, however, acknowledged that thermal decomposition of solutions within process vessels containing flammable solvent can arise from fire engulfment.
4.0 CONSEQUENCE ANALYSIS
Consequence analysis involves estimating the consequences of a scenario and assessing the potential level of harm (or injury) resulting from these consequences.
For this study the level of harm chosen as the basis for assessment is the ‘risk of fatality’.
4.1 Toxic Gas Effects
The object of toxic gas release analysis is to determine whether an adverse health outcome can be expected following a release and if the data permits, to estimate the extent of injury or fatalities that are likely to result. Many useful measures are available to use as benchmarks for predicting the likelihood that a release event will result in injury or death. A.1.Chem.E reviews various toxic effects and discusses the use of various established toxicological criteria (Ref. 1).
For this study the criteria of Immediately Dangerous to Life and Health (IDLH) is utilised. IDLH levels are established by the American National Institute for Occupational Safety and Health and are a common industry standard.
An IDLH exposure condition is defined as a condition “that poses a threat of exposure to airborne contaminants when that exposure is likely to cause death or immediate or delayed permanent adverse health effects or prevent escape@om such an environment “.
IDLH values also take into consideration acute toxic reactions, such as severe eye irritation, that could prevent escape. As a safety margin IDLH values are based on the effects that might occur as a consequence of 30 minute exposure.
Table 1: Toxic Gas Concentrations of Interest
Material IDLH
Nitric Acid 25 wm
Nitrogen Dioxide 20 ppm
For this study the risk of fatality as a result of being exposed to a toxic dose above IDLH is taken as 1%. This probability increases to 100% at the source of the gas release. /*
The consequences and extent of injury that can arise from the toxic gas scenarios as considered in Section 3.1 are assessed through the methodology as defined by the A.1.Chem.E. (Ref. 1). The results are given in Appendix 2 & 3 and are summarised below in Table 2:
The software package PHAST-Professional is utilised to assist in determining these results.
Table 2 - Toxic Gas Releases Consequence Analysis Results
Scenario Description Distance to Receiving an IDLH Dose No. (metres)
la A catastrophic leak from Nitric Acid storage tank TA-640 1 into the tank bund.
Neutral Weather Night Time Weather Conditions (D-5) Conditions (F-1.5)
50 i90
lb A fracture on the 50mm Nitric Acid transfer line from a road tanker to tank TA-6401.
180 710
lc A fracture on the 40mm Nitric Acid transfer line from tank TA-6401 to the Nitration Plant.
125 750
2b The emission of NOx fumes as a result of the catastrophic failure of RE-6 10 1.
4.2 Explosion Scenarios - Blast or Overpressure Effects
An explosion can be defined as “u rapid expansion of gases resulting in a rapidly moving pressure wave or shock wave “.
For a QRA study the consequences of concern are mainly the blast overpressures generated and the projectile effects. The explosion impacts on equipment and people, being of main concern. Blast overpressure estimates can be determined by a variety of models, including the TNT equivalency model, and Prugh’s method for determining overpressure from a ruptured vessel (Ref. l), both models of which are utilised here for consequence analysis.
> The TNT equivalency model is based on the assumption of equivalence between the material under study and TNT.
> Prugh’s method is based on a sphere burst containing an idealised gas. The air shock having its maximum over-pressure at the contact surface between the gas sphere and the air.
4.2.1 Overpressure Consequence Results
4.2.2. I For Scenario 2a the overpressure effects as a result of catastrophic failure of reactor RE-6 10 1 are given in Appendix 4 and are summarised below in Table 3. The consequences and extent of injury that can arise are assessed using Prugh’s model as defined by the A.1.Chem.E. (Ref. 1).
For these results it is assumed that the reactor fails at an over-pressure 4 times the design pressure.
4.2.1.2 For Scenario 3 the consequences and extent of injury that can arise from the detonation of ISDN are assessed using the TNT equivalency model as defined by the A.1.Chem.E. (Ref. 1).
The results are given in Appendix 5 and are also summarised below in Table 3.
As a comparison to the above results Table 4 shows damage estimates for common structures based on over-pressure effects. The interpretation of this data is clear with respect to structural d,amage but subject to debate with respect to human causalities. Some studies equate building damage to a fatal effect, as those inside buildings would probably be crushed.
People outside of buildings or structures are susceptible to:
1. Direct blast injury (blast overpressure).
2. Indirect blast injury (missiles or whole body translation)
Relatively high blast over-pressures >lbarg are necessary to produce direct fatality (primarily due to lung haemorrhage). It is generally believed that fatalities arising from whole body translation are due to head injury from decelerative impact.
To determine the degree of injury from over-pressure effects the following criteria are used to estimate fatality levels due to the direct effect of over-pressure (Ref 3):
For a peak side over pressure of:
l > 0.3 barg the probability of death equates to 1 .O for both people indoors and outdoors.
. > 0.1 barg the probability of death equates to 0.025 for people indoors and 0.0 for people outdoors.
. [ 0.1 barg the probability of death equates to 0.0 for both people indoors and outdoors,
Table 4: Damage Estimates for Common Structures Based on Overpressure.
Probable total destruction of buildings; heavy machine tools (7000 lb) moved and badly damaged; very heavy machine tools (12,000 lb) survive. T :--:A -IT ---Le.. 1:-
Note: The above injury probabilities do not include the effects of missiles, fragments, or shrapnel, which possibly could cause ‘random’ serious injury and property damage at much greater distances.
Page II
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:09
SIFA Ltd, Shannon, Co. Clare.
Quantij?ed Risk Assessment - Nitration Plant - May 2002
5.0 LIKELIHOOD ESTIMATION
Likelihood estimation is the methodology used to estimate the frequency or probability of occurrence of the major hazard scenarios considered in Section 4.0.
For this examination historical data from benchmark studies carried out within the major hazards industry are utilised as follows:
Table 5 - Frequency Failure Iiata
Scenario No. /Description Frequency (/yr) Comment
la. A catastrophic leak from Nitric Acid storage tank TA-6401 into the tank bund. lb. A fracture on the 50mm Nitric Acid transfer line from a road tanker to tank TA-6401.
lc. A fracture on the 40mm Nitric Acid transfer line from tank TA- 6401 to the Nitration Plant.
2a. E-6101 catastrophic failure.
2b. The emission of NOx fumes as a result of the catastrophic failure of RE-6101. 3. ISDN Detonation
5x1C6
4 x 1o-5
4 x 1o-5
5 x 1o-6
5 x 10‘6
1 x 1o-5
Frequency assigned for an instantaneous release from a single contained atmospheric storage vessel. (Ref. 3 - Purple Book). Frequency assigned for a full-bore hose release based on 10 road tanker deliveries per year. Unloading time 1 hour per delivery. (Ref. 3 - Purple Book). Frequency assigned based on 40mm diameter pipeline for a full bore rupture. Total length of pipeline 40m. ’ (Ref. 3 - Purple Book). Frequency assigned based on that given for a reactor vessel. (Ref. 3 - Purple Book). GC ‘L
Frequency assigned based on that quoted for the storage and handling of explosive material. (Ref. 3 - Purple Book).
These frequency values are based on historical data and experience gathered from the operation of other similar types of facility - generally referred to as the implicit approach in risk assessment.
The implicit approach is discussed further in Section 7.2.
The results of the consequence analysis from Section 4.0 are now combined with the frequency estimates from Section 5.0 to provide a measure of risk. In Quantified Risk Assessment a commonly used risk parameter is Individual Risk.
6.1 Individual Risk Profiles
Individual risk is the frequency at which an individual suffers a defined degree of injury.
Individual risk is usually used to indicate how significant the imposition of risk is compared with the background risk an individual is normally exposed to. Risk profiles are plotted on ordnance maps providing an overview of the geographical risk distribution.
6. I. 1 Determination of Individual Risk Profiles
6.1.1.1 Toxic Gas
For toxic gas effects individual risk is calculated taking into consideration:
Pi = probability of fatality at a given location (consequence assessment). Pwd = probability of wind blowing in that direction. Pws = probability of the weather class. fi = frequency of the release scenario (/year).
For each failure event consequence calculations are performed at representative weather conditions for Shannon to establish a good view of possible effect zones. The risk profile of Figure 1 is divided into eight sections (i.e. 1 to 8) representing the directional probability of the wind for that section. The 7 radii represent the distance from the site at 1OOm intervals - ranging from 1OOm through to 700m respectively. The centre point of the map is the Nitration Plant.
The individual risk for a toxic gas release at a given location is evaluated by:
Individual Risk (IR) =&.pwdpwppi = chances offatality per year (Ref 2.)
The individual risk values for the Nitration Plant are shown in Appendix 6 and plotted on Figure 1. The risk contours (or ‘isorisk’ lines) show points of equal risk around the facility.
6.1.1.2 Overpressure Effects
Individual risk for overpressure effects for the reactor vessel failure and that for an ISDN detonation are calculated taking into consideration the following:
= probability of fatality at a given location (consequence assessment). = frequency of the release scenario (/year).
The individual risk at a given location is therefore evaluated by:
Individual Risk (IR) = fi.pi = chances offatality per year (Ref 2.)
Combining the consequence data from Section 4.0 and the likelihood data from Section 5.0 individual risk values for Scenarios 2a, & 3 are given in Table 6 and plotted in Figure 2.
Probability of Death Individual Risk (IR) Distance from (PI 1.;
origin (m) 5 x 1o‘6 5
0.025 1.2 x 1o-7 15 1.0 1 x 1o-5 39
0.025 2.5 x 1O-7 80
Note: The above risk estimates are calculated independently of other site major hazards.
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:09
SIFA Ltd, Shannon, Co. Glare.
QuantiJied Risk Assessment - Nitration Plant - May 2002
7.0 RISK ANALYSIS
7.1 Analysis of the Individual Risk Profile
The individual risk estimates of Section 6:O represent the risk of fatality versus distance arising from the Nitration Plant. The individual risks are plotted connecting points of equal risk of fatality. Values of risk are expressed as chances of fatality per year or yr-’ .
In assessing the individual risk profiles shown of Figures 1 and 2 the following basic principle should be considered:
l The risk from a major hazard to an individual employee, or member of the public should not be significant when compared to other risks to which a person is exposed in everyday life.
l For an employee the maximum tolerable risk at work is typicah’y set at around I in 1000 per annum (i.e. IOe3/j$ (ReJ 5).
l For “a member of the public the maximum tolerable risk from any large scale industrial hazard shouId at least be ten times lower, i.e. IO-“/y.
Note: Public refers to people and areas outside of the site boundary fence for a major hazard installation.
7.1.1 Figure 1 - Toxic Gas Profile
From Figure 1 it can be seen that:
l The risk zone of lo-‘/year falls within a 1OOm metre zone from the Nitration Plant.
o The extent of the risk zone concerned with employee safety (i.e. 10”) therefore falls within this zone and the immediate area of the Nitration Plant/Nitric Acid storage area.
o The extent of the risk zone concerned with public safety also falls within this 1 OOm zone.
This 1OOm zone does however extend partially outside the boundary fence at the North/West side of the SIFA site.
l The risk zone of 10e6/year extends to between 600 and 700m from the site.
Risk zones of between 10-‘/y and 10-“/y have implications for land use planning control as shown in Appendix 7.
7.1.2 Figure 2 - Overpressure/BEast Risk Profile
From Figure 2 it can be seen that:
l The risk zone of 1 OS5 /year falls within a 40m metre zone extending from the Nitration Plant.
This risk zone does not impact outside the boundary fence.
l The extent of the risk concerned with public safety does not extend outside of the site i.e. the 10m7 /year falls within a 1OOm metre zone extending from the Nitration Plant.
7.2 Ho6 are these risks assessed and are they acceptable?
The frequency values utilised in compiling the risk profiles for SIFA are based on historical data and experience gathered from the operation of other similar types of facility - generally referred to as the implicit approach in risk assessment.
The implicit approach relies on the assumption that the history of failures of components includes a wide variety of factors such as design, construction, commissioning, operation, inspection, maintenance and repair, to decommissioning.
Ultimately these factors are representative of the standards of safety in place for the facilities from which the data is gathered.
Failure rates are then applied on the assumption that they represent an ‘average’ value gathered from a cross section of industry, i.e. the data is applied to a facility (i.e. SIFA) on the basis that the standards of both the hardware and that of safety management are considered to be at least of ‘average’ standard.
The Preliminary Hazard Analysis studies conducted as part of this study highlighted a number of safety and environmental issues where further information is required to be compiled and where information management and control is essential to ensuring that the risk of a major hazard scenario being realised is minimised. These issues are relevant to the frequency values utilised in Section 6.0 and that have an influence on the individual risk profile as represented by Figures 1 & 2.
For example, the importance of a formalised process safety management system within SIFA was highlighted by the PHA studies and its role in major hazard control. Some key components are discussed in Section 7.2.1 below.
7.2. I Process Safety Management System
An integral requirement of the Seveso II Directive is to have in place a management system designed to ‘guarantee a high level ofprotection ’ against major hazard scenarios.
At the time of compiling this report a review (and ongoing development) of the process safety management system within SIFA was being undertaken.
Aspects of the process safety management under review include:
7.2.1.1 Chemical Reaction Hazard Studies
Chemical reaction hazard studies set out templates for gathering key data on the chemistry and reaction hazards associated with a process under study. The information serves as a precursor to the design of plant, the further hazard studies to be undertaken, and ultimately defining the basis for the safe operation of the process and equipment under study.
A draft SOP for a chemical reaction hazard assessment has been compiled within SIFA.
7.2.1.2 Hazard and Operability Study
The principle method for conducting process hazard identification and risk assessment studies within SIFA is using the technique of Hazard and Operability Study (HAZOP).
In applying this technique within SIFA a number of proposed changes have being outlined particularly in respect to HAZOP study preparation and HAZOP study follow-up.
For example, ensuring that a mechanism is adopted to identify ail critical safety and environmental equipment and suitable protocols for the maintenance and upkeep of this equipment put in place.
7.2.1.3 Basis of Safety Reports
On completion of a process hazard study and implementation of the action items arising a basis of safety and’environmental protection report should be compiled to:
l And acknowledge closure of all action items raised during the various stages of a process hazard study.
7.2.2 Preliminary Hazard Analysis -Action Items
Regular reviews of the action items highlighted by the Preliminary Hazard Analysis studies should be conducted until the issues identified have been addressed.
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:09
0
SIFA Ltd, Shannon, Co. Glare,
Quantijied Risk Assessment - Nitration Plant - May 2002
8.0 REFERENCES
Ref. 1 ‘Guidelines for Consequence Analysis of Chemical Releases’, Center for Chemical Process Safety of the American Institute of Chemical Engineers, i 999.
Ref. 2 ‘Guidelines for Chemical Process Quantitative Risk Analysis - 2nd Edition’, Center for Chemical Process Safety of the American Institute of Chemical Engineers, 2000.
Ref. 3 ‘Guidance for Quantitative Risk Assessment - Purple Book’, Committee for the Prevention of Disasters, The Director General for Social Affairs and Employment, The Hague, Netherlands, 1999.
Appendix 2 - Scenario la PHAST Dispersion Modelling Data.
SIFA-Nitration Plant
Nitric Acid
Tank TA-6401-Spill Base Case
Material
Scenario
.1
Location
User-Defined Data
Material Identifier NITRIC ACID Type of Vessel Unpressurized ( at atmospheric pressure) Pressure Specification Pressure not used Discharge Temperature 15 c Inventory of material to discharge 3E4 kg
Type of Event Catastrophic rupture Phase Liquid Tank Head Om
Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 0.5 ppm Averaging time associated with Concentration Toxic Status of Dike Area of Dike ERPG selection IDLH selection STEL selection User Defined Averaging
Dike present 30 m2
ERPG is not set IDLH is not set STEL is not set
No user defined averaging time supplied Discharge Parameters
Release height Dispersion
Ignition Location Inventory of material to Disperse
CASE Name: Data
Om
No ignition location 3E4 kg
Discharge Data
User-Defined Quantities Material Temperature Pressure Inventory Scenario Calculated Quantities
NITRIC ACID 15.00 c
1.01 bar 30,OOO.OO kg
Catastrophic Rupture
Page 29
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:09
SIFA Ltd Shannon, Co. Clare.
Quanti$ed Risk Assessment - Nitration Plant - May 2002
Nitration Plant - PHA Study Meeting -Summary Sheet
PHA Date(s): 28.08.01; 06.09.01; 03.10.01.
Area/Project: Nitration Plant.
PHA Description: Processes studied:
l 587 Nitration Solution l 588 Hydrolysed Nitration Solution. . 589 Crude 5-ISM Solution. l 590 Crude 5-ISM.
Scope of PHA:
l 068 Pure 5-ISM. l 555 ISDN. . 840 ISDN.
The following is a summary of the major hazard scenarios identified and associated with the Nitration Plant. Some of
To identify major hazard scenarios associated with the Nitration Plant and prioritise these scenarios for further detailed analysis.
This PHA forms part of an overall site major hazard identification study - arising from EC Directive 96/82 on the control of,major hazard installations.
SUMMARY
these scenarios have common characteristics.
The scenarios are ranked according to the degree of severity (and priority for further analysis) as assigned by the PHA study group and in accordance with the guidelines set out for conducting Preliminary Hazard Analysis.
l Acetyl Nitrate formation - high concentration leads to violent decomposition. Similar consequences obtained where insufficient dilution with Acetic Acid occurs or where excessive lumps of Isosorbide form.
l Potential domino effect arising.
l Temperature > 120°C leads to severe thermal decomposition. Scenario common to a number of process streams within Nitration building.
l Centrifuge of ISDN - impact energy of centrifuge knife on ISDN - potential detonation.
l Failure of high/low water level on Quench Tank TA 6101 potentially leading to catastrophic vessel failure.
* Nitric Acid/Acetic Anhydride - major leak to surface water drain. Potential route leading to environmental impact on local watercourse.
Priority
A
A
A
A
B
B
Page 22
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:09
SIFA Ltd, Shannon, Co. Glare.
QuantiJed Risk Assessment - Nitration Plant - May 2002
Scenario No. 555 /840 -07 555 /840 - 09
Severity 4
l Gas generation from thermal decomposition of process streams e.g. ISDN. Quantity of gas/composition to be determined.
l Transportation of pure ISDN - potential detonation from impact - road transport collision or handling incident.
Priority
B il B
Scenario No. 587 - 06
Severity 3
l Nitric Acid Storage Tank 20,000 litres - major leak into bund or on transfer line.
Priority 3 B
Notes: Some other scenarios were not assigned a sever@ rating until compIetion offurther assessment of hazard study data.
QuantijTed Risk Assessment - Nitration Plant - May 2002
Appendix 2 - Scenario la PJZAST Dispersion Modelling Data.
SIFA-Nitration Plant
Nitric Acid
Tank TA-6401-Spill Base Case
Data SIFA-Nitration Plant\Nitric Acid\Tank TA-6401-Spill Material
Material Identifier NITRIC ACID Type of Vessel Unpressurized ( at atmospheric pressure) Pressure Specification Pressure not used Discharge Temperature 15 c Inventory of material to discharge 3E4 kg
Scenario Type of Event Catastrophic rupture
it Phase Liquid Tank Head Om
Location Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 0.5 ppm Averaging time associated with Concentration Toxic Status of Dike Dike present Area of Dike 30 m2 ERPG selection ERPG is not set IDLH selection IDLH is not set STEL selection STEL is not set User Defined Averaging No user defmed averaging time supplied
Discharge Parameters Release height Om
Dispersion Ignition Location No ignition location Inventory of material to Disperse 3E4 kg
Appendix 2 - Scenario la PHAST Dispersion Modelling Data.
SIFA-Nitration Plant
Nitric Acid
Tank TA-6401-Spill Base Case
Material
Scenario
.1
Location
User-Defined Data
Material Identifier NITRIC ACID Type of Vessel Unpressurized ( at atmospheric pressure) Pressure Specification Pressure not used Discharge Temperature 15 c Inventory of material to discharge 3E4 kg
Type of Event Catastrophic rupture Phase Liquid Tank Head Om
Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 0.5 ppm Averaging time associated with Concentration Toxic Status of Dike Area of Dike ERPG selection IDLH selection STEL selection User Defined Averaging
Dike present 30 m2
ERPG is not set IDLH is not set STEL is not set
No user defined averaging time supplied Discharge Parameters
Release height Dispersion
Ignition Location Inventory of material to Disperse
CASE Name: Data
Om
No ignition location 3E4 kg
Discharge Data
User-Defined Quantities Material Temperature Pressure Inventory Scenario Calculated Quantities
Data SIFA-Nitration Plant\Nitric Acid\SOmm-RoadTanker-Spill Material
Material Identifier NITRIC ACID Type of Vessel Unpressurized (at atmospheric pressure) Pressure Specification Pressure not used Discharge Temperature 15 c Inventory of material to discharge 2E4 kg
Scenario Type of Event Line leak
.* Phase Liquid Pipe Diameter 50 mm Supply Pump Head No Tank Head 3.5 m Number of Excess Flow Valves 0 Number of Non-Return Valves 0 Number of Shut-Off Valves 0
Location Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 0.5 ppm Averaging time associated with Concentration Toxic Status of Dike Dike present Area of Dike 30 m2 ERPG selection ERPG is not set IDLH selection IDLH is not set STEL selection STEL is not set User Defined Averaging No user defined averaging time supplied
Indoor/Outdoor Outdoor Release Direction Horizontal
Discharge Parameters Default line length 0.1 m Release height 0.5 m
Dispersion Ignition Location No ignition location Inventory of material to Disperse 2E4 kg
Appendix 2 - Scenario 1 b PHAST Dispersion Modelling Data.
SIFA-Nitration Plant
Nitric Acid
SOmm-RoadTanker-Spill Base Case
User-Defined Data
Material Material Identifier NITRIC ACID Type of Vessel Unpressurized ( at atmospheric pressure) Pressure Specification Pressure not used Discharge Temperature 15 c Inventory of material to discharge 2E4 kg
Scenario Type of Event Line leak
,. Phase Liquid Pipe Diameter 50 mm Supply Pump Head No Tank Head 3.5 m Number of Excess Flow Valves 0 Number of Non-Return Valves 0 Number of Shut-Off Valves 0
Location Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 0.5 ppm Averaging time associated with Concentration Toxic Status of Dike Area of Dike ERPG selection IDLH selection STEL selection User Defined Averaging
Indoor/Outdoor Outdoor Release Direction
Discharge Parameters Default line length Release height
Dispersion Ignition Location Inventory of material to Disperse
Data SIFA-Nitration Plant\Nitric Acid\40mm-Transfer-Line Material
Material Identifier Vessel
Release Type Location
Northern location of dispersion source Eastern location of dispersion source Dispersion Concentration of Interest Averaging time associated with Concentration Status of Dike
r* ERPG selection IDLH selection STEL selection User Defined Averaging
Indoor/Outdoor Outdoor Release Direction
Discharge Parameters Release height
Dispersion Number of Release Segments Fluid Phase(l) Discharge Velocity( 1) Droplet Diameter(l) Duration of Discharge( 1) Final Temperature( 1) Liquid Fraction( 1) Release Rate( 1) Ignition Location Inventory of material to Disperse
NITRIC ACID
Om Om
0.5 ppm Toxic
No dike present ERPG is not set IDLH is not set STEL is not set
No user defined averaging time supplied
Horizontal
10.5 m
1 Liquid
1.1 m/s 1000 mm 600 s
15 c 1 fraction
2.11 kg/s No ignition location
3E4 kg
Page 3 7
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:10
SIFA Lfd, Shannon, Co. Clare.
QuantiJed Risk Assessment - Nitration Plant - May 2002
Location Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 0.5 ppm
. . Averaging time associated with Concentration Toxic No dike present ERPG is not set IDLH is not set STEL is not set
No user defined averaging time supplied
Status-of Dike ERPG selection IDLH selection STEL selection User Defined Averaging
Indoor/Outdoor Outdoor Release Direction
Discharge Parameters Release height
Dispersion Number of Release Segments Fluid Phase( 1) Discharge Velocity( 1) Droplet Diameter( 1) Duration of Discharge( 1) Final Temperature( 1) Liquid Fraction(l) Release Rate( 1) Ignition Location Inventory of material to Disperse
CASE Name: Data
Horizontal
10.5 m
1 Liquid
1.1 m/s 1000 mm 600 s
15 c 1 fraction
2.11 kg/s No ignition location
3E4 kg
Page 38
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:10
SIFA Ltd, Shannon, Co. Glare.
QuantiJed Risk Assessment - Nitration Plant - May 2002
Data SIFA-Nitration Plant\Nitric Acid\RE-6101~NO2 Material
Material Identifier NITROGEN DIOXIDE Vessel
Release Type Instantaneous Location
Northern location of dispersion source Otll Eastern location of dispersion source Om Dispersion Concentration of Interest 1 Pw I. Averaging time associated with Concentration Toxic
No dike present ERPG is not set IDLH is not set STEL is not set
No user defined averaging time supplied
Status of Dike , ERPG selection IDLH selection STEL selection User Defined Averaging
Discharge Parameters Release height
Dispersion Number of Release Segments Fluid Phase( 1) Discharge Velocity( 1) Final Temperature( 1) Liquid Fraction(l) Ignition Location Inventory of material to Disperse
10 m
1 Vapour
50 m/s 120 c
0 fraction No ignition location
57 kg
Page 43
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:10
SIFA Ltd, Shannon, Co. Clare.
Quantified Risk Assessment - Nitration Plant - May 2002
Location Northern location of dispersion source Om Eastern location of dispersion source Om Dispersion Concentration of Interest 1 wm Averaging time associated with Concentration Toxic .1 Status of Dike No dike present ERPG selection ERPG is not set 1DLH selection IDLH is not set STEL selection STEL is not set User Defined Averaging No user defined averaging time supplied
Discharge Parameters Release height 10 m
Dispersion Number of Release Segments 1 Fluid Phase( 1) Vapour Discharge Velocity( 1) 50 m/s Final Temperature( 1) 120 c Liquid Fraction( 1) 0 fraction Ignition Location No ignition location Inventory of material to Disperse 57 kg
CASE Name: Data
Consequence Results
Distance to Concentration Results Concentration(ppm) Averaging Time Distance (m)
C 9.85 9.85 9.85 C 9.85 9.85 9.85 fraction 0.7 0.7 0.7
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:11
l
SIFA Ltd, Shannon, Co. Clare.
Quantijied Risk Assessment - Nitration Plant - May 2002
Appendix 4 Blast Effects
Consequence Estimates for Failure of Reactor RE6101.
I. Scenario 2a.
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:11
SIFA Ltd, Shannon, Co. Glare.
Quantijied Risk Assessment - Nitration Plant - May 2002
SIFA Shannon - QRA Study: March 2002. Scenario 2a: Nitration Plant RE 6101
Input Data: Vessel burst pressure: Distance from vessel center: Vessel vapour volume: Final pressure: Heat capacity ratio: Molecular weight of gas: Gas temperature:
12 bar abs 15m 2.3 m**3
I .01325 bar abs 1.31
46 393 K
Design Pressure = 3 bar
Calculated Results: English units equivalents of above data:
Vessel burst pressure: Vessel volume:
‘* Final pressure:
174.09 psia 81.21 ft**3
14.7 psia
Energy of Explosion from Brown’s Equation:
Trial and error solution to determine surface pressure via equation (2.2.14): Guessed Value: 3 bar abs Calculated Value: 3.038542 bar abs English Equivalent: 43.52 psia
c--- Adjust until equal to
Trial and error solution to determine virtual distance: TNT Mass: 1.47 kg
Distance from blast: 2.216 m e- Adjust to match surfa above
Quantified Risk Assessment -Nitration Plant - May 2002
Appendix 7 Land Use Planning Controls
For
insp
ectio
n pur
pose
s only
.
Conse
nt of
copy
right
owne
r req
uired
for a
ny ot
her u
se.
EPA Export 25-07-2013:15:02:11
SIFA Ltd, Shannon, Co. Glare.
Quantijied Risk Assessment - Nitration Plant - May 2002
Appendix 7 - Guidance on land use planning under by EC Council Directive 96f82
Category of Development A. Housing, hotel, holiday accommodation.
B. Workplaces, Parking areas.
C. Retail outlets, Community and leisure facilities!’
D. Institutional establishments and special accommodation.
Development Type and Size
EXCLUSIONS:
1) Accommodation providing for either less than 3 dwelling units or less than 10 people - treat as Category B unless it may set a precedent for substantial further development or will be closer to the hazardous installation than existing development of a similar nature, in which case treat as Category C.
2) Housing accommodation specifically for the elderly or handicapped, e.g. sheltered housing - treat as category D.
3) Accommodation 5 storeys or more in height - treat as category D. Parking areas must be for fewer than 200 vehicles (if larger treat as Category C), and may not have other associated facilities (other than toilets).
EXCLUSIONS: Commercial or industrial development providing for 100 or more occupants or 3 or more storeys in height; commercial and industrial development specifically for the handicapped (e.g. sheltered workshops) - treat both as Category C.
EXCLUSIONS:
1) Retail development with less than 25Om2 gross floor space; community and leisure facilities with less than lOOm2 gross floor space - treat both as category B.
2) Developments with SOOOm2 or more gross floor space -treat as Category D.
3) Predominantly open-air developments where there will be frequently (once a week or more) very large numbers (lOOO+) of people out-of-doors (such as a sports stadium or a large retail market) - treat as Category D.
EXCLUSIONS: Educational establishments where persons under 16 years old will not be present - treat as Category C.
Guidance on risk levels and developments permissible.
Caiegory of development
Inner zone Middle zone Individual risk Individual risk exceeds I@ exceeak 106
A. Housing, hotel, holiday accommodation
B. Workplaces, Parking areas
C. Retail outlets, community and leisure facilities.
D. Institutional establishments and special accommodation